English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The performance of machine learning models often degrades after deploymentdue to data distribution shifts. In many use cases, it is impossible tocalculate the post-deployment performance because labels are unavailable orsignificantly delayed. Proxy methods for evaluating model performancestability, like drift detection techniques, do not properly quantify datadistribution shift impact. As a solution, we propose a robust and accurateperformance estimation method for evaluating ML classification models onunlabeled data that accurately quantifies the impact of covariate shift onmodel performance. We call it multi-calibrated confidence-based performanceestimation (M-CBPE). It is model and data-type agnostic and works for anyperformance metric. It does not require access to the monitored model - it usesthe model predictions and probability estimates. M-CBPE does not need userinput on the nature of the covariate shift as it fully learns from the data. Weevaluate it with over 600 dataset-model pairs from US census data and compareit with multiple benchmarks using several evaluation metrics. Results show thatM-CBPE is the best method to estimate the performance of classification modelsin any evaluation context.

We don't need no labels: Estimating post-deployment model performance under covariate shift without ground truth